Manipulator Control by Visual Servoing with Stereo Vision

Abstract

This paper presents a new method of visual servoing with stereo vision to control the position of a manipulator with respect to an object. Conventional control methods by visual servoing use a monocular camera and have several problems. For example, either shape information or desired distance of the target object from the camera must be given. Furthermore, it is not stable if the initial positional error of features in the image is very large. These problems are caused by image the Jacobian matrix, that is, the approximate value at desired position is used instead of correct one. By using stereo vision, the image Jacobian matrix can be calculated correctly at any position. So neither shape information nor desired distance of the target object is required. It is also stable even if the initial error is very large. Both simulation and experimental results are presented to demonstrate the effectiveness of this method.